I'm also excited about this course like never before. Although, Udacity already has courses on machine learning that cover everything from the basics to reinfor

I'm also excited about this course like never before. Although, Udacity already has courses on machine learning that cover everything from the basics to reinforcement learning and Tensor Flow, this course is much more hands-on and will probably cover a wider range of topics like fusing data from sensors and algorithms for localization and mapping the environment. I studied computer science but remained interested in electronics, mechanics and signals and system. One of the projects I did was a GPS guided rover. I did all the electronics, programming and mechanics stuff and wrote basic obstacle avoidance, localization and navigation code myself. I enjoyed every part of it. I wanted to make the rover very sophisticated, which retrospectively was a very high goal for a semester-long course. Anyway, I went ahead and bought two books, Kalman Filtering: Theory and Practice Using MATLAB and Principles of Robot Motion, co-authored by Sebastian Thrun. I didn't get the time to read beyond a few chapters of Thrun's book then but it was very well written and included some great stuff like an introductory chapter on topology. Eight years since and it seems finally I have the motivation to read it cover to cover. Since the course starts later this year there's time to learn Deep Learning and perhaps Gazebo simulator, which we might end up using for simulating of our own self-driving cars.